|Theme||Visible||Selectable||Appearance||Zoom Range (now: 0)|
In the past, much of the petrophysics done in the Australian mining industry has been based upon gamma ray, simple density devices, resistivity, and televiewers. Common uses of petrophysical data include locating the top and bottom of the seam/ore, determining the water level, mapping fractures and faults, computing hardness, and facies analysis. However, the industry is moving toward more advanced applications, such as improved methods of understanding the porosity and permeability of the rocks, 3D mapping of stability, and the use of petrophysical measurements as a cost-effective means of supplementing or even replacing traditional assay methods.
This paper begins with a brief introduction to the mining environment as compared with the modern oilfield environment. While petrophysical data acquisition in East Australian coal mines is not so far removed from shallow oilfield land wells, open pit mines, such as the Pilbara Iron Ore fields of Western Australia are a very different world - thousands of holes are drilled, each generally less than 60 metres. Assays (geological analysis of material collected from the hole) are the primary reference data. Costs to log are low and many processes (data interpretation, delivery of logs, etc.) are automated.
Next we will review how gamma ray, density, neutron, resistivity, and caliper measurements are used throughout the Australian mining industry, paying some attention to the challenges of using classic tool designs such as 16/64 normal resistivity tools and single point (uncompensated) density. Sonic, electrical imaging, and optical televiewers are the next tier of measurements, used for fracture/fault mapping, ground stability, hardness and seismic integration. Finally, we will discuss the latest wave of technologies to be gaining ground in the Australian mining market, including NMR, VSP, and elemental spectroscopy.
The introduction of advanced petrophysical measurements in Australian mining is opening the door for exploiting new applications, many centered around “big data” or machine learning techniques, such as automated facies identification, high resolution mapping of both major and minor minerals, and 3D visualisation of ore properties.
Horizontal wellbores are challenging environments in which to understand borehole acoustics measurements, particularly in shale reservoirs. However, with careful analysis, the data offers a wealth of geomechanical and rock physics information not only about the formation in which the tool is located, but about nearby formations as well. This not only allows for production optimization by understanding how fractures will propagate, but also suggests geosteering opportunities, such as placing the lateral a safer distance from nearby boundaries, or steering the well to stay within the most “fracable” portion of the shale.
A case study is presented, considering azimuthal acoustic data that was acquired in a thinly-bedded formation. The target formation is heterogeneous and thinly-bedded with highly-contrasting rocks acting as both source and reservoir in the hydrocarbon petroleum system. In this particularly challenging example, the borehole straddled the interface of two highly contrastive beds for much of the length of the lateral. Initial sonic interpretation using conventional methods led to confusing results; it was difficult to understand whether the velocities pertained to the upper formation, lower formation, or beds even further away.
A new workflow was applied to try to systematically characterize the sonic properties for each layer. This involves visualizing the environment, determining the sonic velocities for each quadrant (up, down, left, right), and using the visualization to determine which velocities (and thus elastic properties) belong to each layer. Once elastic properties for each layer were understood, it is possible to try to solve the problem of how the complex fracture propagation should be modelled in these challenging environments. When further combined with natural fracture information (from image logs or sonic fracture detection), core measurements, and past production from nearby wells, these analyses can be used as a guide for modifying and optimizing completions.
The stratigraphy of the Delaware Wolfcamp is unique amongst “unconventional” plays. It is thinly bedded, composed of mixed fine grain carbonates (siliciclastic lithotypes) and exhibits highly contrastive “stiff” and “soft” facies (Thiercelin and Plumb, 1994).
Acoustic anisotropy measurements are used in a wide variety of applications, such as fracture characterisation, production optimisation, wellbore stability, and hydraulic fracture monitoring. However, the translation of acoustic waveforms to anisotropic velocities and stress directions is often seen as a “black art”. While the acoustics specialist knows just how factors such as flexural mode dispersion, drilling induced stress, high angle fractures, hole washout, centralisation and tool calibration affect the data, the end user generally doesn’t want to delve deeply into such details, but just wants to know if the results are valid for a particular application.
Thus, while this paper will begin with a review of the common industry methods of processing acoustic anisotropy, the main focus will be on the quality control and understanding how to identify common issues in the delivered results. To that end, a review of quality control plots and their interpretation will be presented, followed by numerous examples of valid and invalid processed results, with the goal of this work being not to instruct the user on how to fix problem datasets, but rather to detail a succinct workflow to identify when to trust anisotropy results and when to consider re-processing the waveform data.
Particular attention is paid to frequently asked questions, such as, “what is the difference between XX/YY processing and shear anisotropy analysis?”, “how do I interpret anisotropy in inclined and horizontal wells?”, “how off-centre can the tool be to get valid shear anisotropy results?”, “what does a dispersion crossover plot really mean?” and “can I get anisotropy from LWD?”
Acoustic anisotropy information is needed for many applications, yet the process for determining anisotropy and associated quality control methods are often poorly understood by the end user. This can cause users to incorporate incorrect results in various applications, leading to a lack of confidence in acoustic data. While there are numerous ways of computing and delivering anisotropy data and accompanying quality control plots, there are relatively simple and practical methods that the end user can practice to determine whether 1) the processed results are valid representations of formation anisotropy (rather than tool or environmental effects) and 2) what those results mean.
This paper is targeted at the non-specialist in acoustics and aims to provide a guide to understanding processed acoustic anisotropy results and to recognize valid and invalid data. It is not intended as a manual on how to process anisotropy; rather it addresses frequently asked questions for the practical user.
Acoustic data has long been used to characterise fractures, but the methods are often misunderstood, particularly in inclined wells, unconventional reservoirs, and through casing. For example, azimuthal shear anisotropy can be an excellent indicator of fractures, but not if the fractures are perpendicular to the wellbore, which is often the case in horizontal wells. However, Stoneley fracture detection and amplitude/attenuation methods are suited to detect fractures transecting the wellbore. While, in general, the preferred method to characterize near-wellbore fractures is to use electrical or ultrasonic image tools, sonic data offers not only an alternative for identifying fractures when image logs cannot be acquired, but also complement image logs by reading both near and far from the wellbore as well as identifying stress concentrations which are likely to fracture when the stress field is altered by drilling and production. Zones which are stressed (and prone to fracturing) but have not yet fractured are not readily identified with micro-imagers, but can be identified by acoustic logs. Sonic logging also offers possibilities for fracture characterization behind casing, which is not possible with micro-imagers. Acoustic methods also work in all mud types (except air). Much like electrical or ultrasonic imaging methods, there are environments in which sonic fracture characterization techniques work well and those where they are not ideally suited.
This paper begins with a review of acoustic fracture characterization methods, which environments each method is best suited for, and suggesting optimal environments and limitations of the measurements. Examples of the various techniques will be discussed, comparing with non-acoustics methods to understand how different methods complement one another.
Understanding natural and drilling induced fractures is essential to safely drilling a well and maximizing production. While seismic interpretation methods can give us a high level view of areas of fractures, it is at the borehole level we can really understand how fractures intersecting or near the wellbore may interact. By measuring the discrete location of fractures intersecting the wellbore, induced fractures, and altered near-wellbore stress fields, we can understand the effect of the wellbore on the surrounding environment. This understanding is critical to safely drilling a stable wellbore and to optimising production, particularly in tight or shale reservoirs.
Acoustic velocities are one of the key measurements for production enhancement and increased drilling efficiency in unconventional reservoirs, but the environment can present considerable complications to accurate data interpretation. Because these reservoirs are often produced by way of horizontal boreholes through shales, the sonic logs respond to intrinsic shale anisotropy, natural fractures, drilling induced stress, and nearby beds above or below the target formation. These intertwined effects frequently lead to misinterpretations such as picking a compressional slowness from an approaching bed but the shear slowness from the current formation, leading to odd (and confusing) geomechanical results. Other times, shale anisotropy is inadvertently assumed to be linked to natural fracture networks. Interpretations from different vendors, and even different analysts within the same team might vary greatly due to different processing methods. Thus, we need to consider the various aspects of sonic response in horizontal wellbore through unconventional reservoirs and proscribe a workflow and quality control regime to ensure consistent and valid results.
We begin with a discussion of the unconventional reservoir applications which require sonic data as key inputs, followed by a brief review of 3D anisotropy. The remainder of the work focuses on understanding the complex sonic log responses in horizontal shale reservoirs. A workflow is described, with the goal of correctly determining the compressional and shear velocities for the bed in which the tool resides as well as the nearby beds. Several field examples will be examined in detail, describing the correct methods for interpretation and the common pitfalls. Examples of how incorrectly interpreted results impact formation evaluation calculations follow. Finally, operational recommendations to optimise sonic data collection in shale reservoirs will round out the discussion.
Ahmad Nazlan, Mohd Firdaus (ADCO) | Raza, Taufique Ahmad (ADCO) | Al Abd Salem, Salem (ADCO) | Al-Amiri, Abdelhameed A (ADCO) | Hafez, Hafez H. (ADCO) | Sallam, Yassin Farouk (ADCO) | Al Arfi, Saif Ali (ADCO) | Mostafa, Hassan (Weatherford) | Market, Jennifer (Weatherford) | Farouk, Omar El (Weatherford)
Landing a high-angle well in the very thin carbonate reservoir of the lower Cretaceous can be challenging due to heterogeneity present in thin layers above the reservoir and lack of correlatable non-porosity measurements. The use of radioactive source porosity is effective, but due to the extended trajectory and high differential pressure in over burden depleted reservoirs, the chance of becoming stuck is a considerable risk and a big concern, and thus the use of classic nuclear porosity tools are untenable. This paper is a case study illustrating the use of azimuthal sonic porosity as a highly effective method of landing the well in the reservoir. Azimuthal sonic data were acquired with a focused unipole tool which recorded the measured waveforms and computed compressional and shear velocities in 16 azimuthal bins. Real-time compressional and shear slownesses were also computed by stacking all 16 bins of data, which gave very high signal-to-noise ratios and excellent data quality - often a challenge in hard formations. These azimuthally averaged slownesses were used by the geologist to identify formation tops while drilling, and to detect the approach of the target reservoir, resulting in a safe landing in the target zone. While the azimuthally averaged real-time sonic porosities were effectively used to land in the target reservoir, the azimuthal results were examined shortly after the well was drilled to understand the additional advantages of using the azimuthal sonic porosities real-time to detect approaching beds from further away. It is clear from the results that, if the porosities from each quadrant (or even just up and down) were transmitted real-time, the reservoir could be detected an additional 1.5 ft.
Acoustic anisotropy analysis is used in a wide variety of applications, such as fracture characterization, wellbore stability, production enhancement, and geosteering. However, acoustic anisotropy methods are not always understood clearly, both by the end user and the data analyst. Azimuthal variations in velocities may be due to stress variations, intrinsic anisotropy, bed boundaries, or some combination thereof. Environmental effects, such as hole inclination, centralization, wellbore condition, dispersion and source/receiver matching, affect the viability of the data to detect anisotropy and must be considered in the interpretation. Untangling the various acoustic anisotropy types from environmental effects is essential to interpret the results correctly.
This paper begins with a discussion of the types of acoustic anisotropy, followed by a review of common industry methods for extracting anisotropy from wireline and LWD azimuthal sonic data. Environmental factors such as tool centralization, irregular borehole shape, poor tool calibration, and dispersion are considered, paying particular attention to the practical limitations of acquiring data suitable for high-quality anisotropy analysis in adverse conditions.
Quality control techniques are discussed in some detail, as they help identify various causes of “false anisotropy” due to processing artifacts. Quality control plots, such as shear slowness images, dispersion crossover plots, and combined analysis with calliper and microimages, are suggested to aid the nonspecialist in determining whether the anisotropy results are viable.
Intrinsic, induced, and geometric anisotropy are discussed in detail, along with consideration of the depth of sensitivity of acoustic measurements. A case study is presented to illustrate the art of untangling overlapping acoustic anisotropy responses.
Acoustic anisotropy analysis is used in a wide variety of applications, such as fracture characterisation, wellbore stability, production enhancement, and geosteering. However, the methods by which acoustic anisotropy are determined are not always well understood, both by the end user and the data analyst. Azimuthal variations in velocities may be due to stress variations, intrinsic anisotropy, bed boundaries, or some combination thereof. Environmental effects such as hole inclination, centralization, wellbore condition, dispersion and source/receiver matching affect the viability of the data and must be considered in the interpretation. Untangling the various acoustic anisotropy factors is essential to effectively interpreting the results.
This paper begins with a discussion of the types of acoustic anisotropy, followed by a review of common industry methods for extracting anisotropy from wireline and LWD azimuthal sonic data. Environmental factors such as tool centralization, irregular borehole shape, poor tool calibration, and dispersion are considered, paying particular attention to the practical limitations of acquiring data suitable for high quality anisotropy analysis in adverse conditions.
Quality control techniques are discussed in some detail, as there are various causes of “false anisotropy” that should be recognized so as not to incorrectly interpret processing artefacts as formation features. Quality control plots are suggested to aid the non-specialist in determining whether the anisotropy results are viable.
Intrinsic, induced, and geometric anisotropy are discussed in detail, along with consideration of the depth if sensitivity of acoustic measurements. Finally, a case study is presented to illustrate the art of untangling overlapping acoustic anisotropy responses.
Borehole sonic data are acquired for a myriad of uses, including pore pressure prediction, porosity, seismic correlation, wellbore stability, hole size determination, fracture characterisation, permeability, cement bond analysis and more. One of the difficulties in the well planning stage is determining the best sonic tool/service for the purpose, as we must balance such factors as cost, data quality, environmental challenges, conveyance risk and timeliness of data delivery to select the right service for the application.
This paper will review a variety of applications and environments, such as the acquisition of compressional data in large shallow wells for accurate pore pressure prediction, obtaining good quality shear data in slow and fast formations, and logging azimuthal shear velocities for production enhancement in unconventional wells. Recommendations will be suggested for optimal and cost-effective sonic logging programmes. LWD and wireline options will be considered as well as a variety of types of hardware. As this is a non-partisan presentation, no trade names or specific services will be referenced. Particular configurations, such as optimal source-receiver spacing, azimuthal capabilities, or transmitter frequency ranges will also be suggested, but again in a generic sense.
While it is natural to focus attention on hardware configuration and signal quality, it is equally important to consider service delivery factors such as reliability of hardware, telemetry speed, waveform processing quality and data delivery. We will consider a variety of scenarios and applications, considering how to weigh these service quality and delivery factors in each instance. For example, if the primary application of sonic data for a well is real-time pore pressure prediction, then telemetry rates, automated processing quality, and data monitoring are critical, but whether the hardware is capable of acquiring high resolution azimuthal data or very slow shear may not be as essential. On the other hand, if the primary application is fracture characterisation in a reservoir, perhaps the azimuthal shear data quality is paramount while the turn-around time for processing the data may be less important. We will recommend standard deliverables for each application.
LWD Sonic logging has made great strides in the last decade, but measuring acoustic anisotropy has been largely out with the abilities of any of the industry tools. We introduce a new crossed-dipole, azimuthal imaging tool capable of measuring anisotropy in the same manner as wireline sonic tools do today, and even surpassing the resolution of wireline tools by taking advantage of the rotational behaviour of LWD tools. When we consider the real-time nature of LWD logging, new applications become possible , such as sonic geosteering, real-time sonic fracture characterisation (which measures deeper than current electrical fracture imaging methods), and azimuthal stress profilingfor optimising wellbore stability.
The tool has 4 azimuthal transmitters which can be operated as a monopole, dipole, crossed-dipole, or quad-rupole as well as 4 azimuthal receiver arrays. An acoustic calliper is also integrated into each of the 4 receiver arrays for accurate measurement of hole size, shape, and tool position.
The physics of LWD sonic logging is, in many ways, more complex than the wireline environment, due to the presence of a substantial drill collar, tool rotation, mud circulation, and drilling noise. We will review the effects of these factors on LWD Sonic data, including dispersion corrections that account for the drill collar and tool position as well as design considerations to minimise the effects of mud circulation and drilling noise. We then show modelling and field data to compare the accuracy of crossed-dipole anisotropy calculations and rotational anisotropy methods. Following that, we will consider the application of sonic imaging for geosteering by way of modelling and field examples